MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Ignition issue

78 Roadster - accuspark ignition - won't start - new coil/leads - new ignition relay - 12volt on the coil with the ignition on but zero volts during cranking - single spark when the ignition is turned off = stumped !
Your thoughts gratefully considered.
Roger
Roger Walker

Hi,

First point, you shouldn't be showing 12 volts at the coil on a rb car as it's a ballast system! More like 9 volts.


It ought to show cranking voltage when turning over, usually 10'5/11 volts.


If you're getting nothing, firstly check the wire hasn't dropped off the small terminal on the starter solenoid, as that bypasses the ballast resistor wire in the loom on cranking.


Try running a wire from the fusebox (white wire) to the coil + terminal, if it then starts you deffo have a problem in the supply to the coil.



The other possibility is a faulty ignition switch or a poor connection, feel around the white wires for terminals getting hot.


Good luck with it.


SR Smith 1

Replace the Accuspark with a set of points. The first thing I would consider is that the Accuspark module could have failed.
Mike Howlett

Accuspark.................rubbish in my opinion! That's the first point.
Second You are obviously only getting any voltage at the coil on "run", not on cranking.
Many later MGB's have had their ballast wire by-passed, but you need to check out all the wiring relating to the ignition system. Also check which coil you have. Put a meter across the primary connections with the wires off. 3 ohms = 12v; 1.5 ohms =- ballast 9v coil. Less than 1.5 ohms is an electronic coil. Then follow Mr Smith's advice.
Allan Reeling

"First point, you shouldn't be showing 12 volts at the coil on a rb car as it's a ballast system! More like 9 volts."

This incorrect on two counts:

1. You *will* see 12v on the +ve terminal of the coil on a ballasted system, unless the points or electronic trigger are effectively 'closed' and so drawing current through the coil.

2. When the points are trigger *are* closed, you should see about 6v on the coil +ve, not 9v, as the coil primary and the external ballast should both measure about 1.5v and so the ignition voltage is split between them equally. This makes the coil a 6v coil, not a 9v coil. You may well see about 9v when the engine is running, but that is because the voltage is switching between 7v and 14v as the points or trigger open and close. Some electronic triggers will show closer to 14v at idle, reducing towards 7v as the revs are increased, and these are variable dwell triggers.

Further to 2 the normal working of the ballasted system means that the ballast is bypassed during cranking connecting full battery voltage to the coil +ve. However due to the load of the starter cranking an engine with plugs fitted the battery voltage - and hence the voltage at the coil +ve - will drop to about 10v with a good battery, less with a weak battery or bad connections in the starter circuit.

Even if the ballast bypass has failed you should still see some voltage between 6v and 10v on the coil +ve, depending on whether you have points and what type of electronic trigger, and only in cases of extreme conditions (or some other fault) should the engine fail to start.

If the points or electronic trigger are faulty at the very least you would still expect to see about 5v on the coil +ve. For the voltage to drop to zero there has to be a failure in the ignition supply i.e. ignition switch or relay - *and* the ballast bypass circuit has to be faulty as the voltage from that comes direct from the solenoid and the battery cable.

1978 was a change point for ignition wiring: In 1977 power for the coil and all the ignition powered electrics came from the ignition relay. During 1978 this changed so that the coil and some ignition powered stuff like the indicators and heater fan was moved back to the ignition switch, but the rest of the ignition powered stuff including the ignition warning light and the main fusebox white/brown to green still came from the relay.

If your car has one in-line fuse going from white/brown to green it has the 77 wiring, if it has two it has the 78 and later wiring. It should be possible to work out if it is the ignition switch that is failing, or the ignition relay, by testing the components I have mentioned during cranking.
Paul Hunt

Thank you all for your comment here. This is a holding response because the car is 50 miles away and I will not be able to get my hands on it until Saturday (pray for good weather).

The first unknown is the car itself; the wiring has been "modified" by a PO and I am not even sure that it still has its ballast resistor (Allan's point). It will only be ballasted if the resistor is built into the wiring as with the V8 because I did not see a resistor on the inner wing (and I did not get to the wiring diagram to see that it was supposed to have one until I got home). However, I am pretty sure it is a 12V coil (but will check).

We checked out the ignition switch and I tried unsuccessfully to reach the starter solenoid to see if the white and light green line was properly attached - that is first on the list on Saturday (SRS) - I shall also investigate whether the car had 77 or 78 ignition Paul.

Allan, I am sure you have good reason to think accuspark is rubbish. Is this a reliability issue, design fault or quality problem ? We don't want to find it has packed up on a stormy night in the middle of nowhere. The car is regular all weather transport.

Thanks again

Roger
Roger Walker

Roger,
Have fitted, or tried to fit, one or two. If they work "out of the box" then they seem OK, unfortunately they often don't work out of the box. To my mind a product which is assembled with such variable outcomes is hardly likely to be the most reliable. Always carry your points and condenser!
MG never used anything other than a resistor wire in the loom. It's output gives the running voltage to suit the lower voltage coil, cranking puts available battery voltage through that coil which gives good coil output to compensate for battery drain while the starter works to spin the engine.
Allan Reeling

In fact the ballast system more than compensates for the load of cranking, as it operates the coil at typically 10v instead of the 6v of normal running.
Paul Hunt

Allan/Paul - thanks, the facts speak for themselves. We have used Lumenition since they came on the market - and we converted the V8 to a later distributor with its own amplifyer. Even then, we carry a spare.
Roger Walker

Well she is running. Disconnected the white and light green line from the starter solenoid and she went straightaway. It seems the starter was grounding the line from the coil when the solenoid was engaged. Is there another explanation?

Roger Walker

If it's a ballasted coil system, there is a switch/relay in the solenoid that bypasses the ballast resister (or engages the ballast resister, I can't remember which but I think bypass is right) when the solenoid is engaged. That switch could be failing and shorting to ground. Jud
J. K. Chapin

That does seem reasonable in theory, but in practice I find it quite surprising.

Having had a problem with the boost circuit in the past (not boosting) I dismantled the solenoid to find and fix the problem. The system on my starter at least is that as the solenoid is activated it moves a copper bar to bridge two studs - one from the battery cable and the other to the starter (A and B in the attached) to spin the starter. The boost contact is a third lightly sprung contact situated between the two studs, so the copper bar bears on that at the same time, and that shorts-out aka bypasses the ballast resistance to put full battery voltage on the coil +ve.

All may be revealed when you dismantle, but the bit I find surprising is that somehow movement of the copper bar is earthing the boost contact, while not earthing the the copper bar itself.

In my case the boost contact (arrowed in the attached) was bent back so the bar missed it altogether, and obviously had never worked. I suppose it's possible that your contact is so badly positioned that some other part of the solenoid push-rod - i.e. an earthed part - is touching it.


Paul Hunt

Rather than a solenoid problem earthing out the coil +, is it not more likely that the fault is on the w/lg wire. Easy to check by running a new wire from the solenoid to the coil+.
Allan Reeling

Jud/Paul/Allan, Thank you. I have of course checked the white and light green wire which is sound. Paul you are right; we shall only find the fault when the solenoid is dismantled -- which won't be for a while because the car now starts very easily and I shan't get my hands on it for a while (Back to the V8). The think that annoys me is the starter was new last summer and starting became an intermittent problem pretty quickly afterwards.
I must remember to tell myself that this car has run for decades without any probs at all - now it has become interesting !
Roger
Roger Walker

If the fault was with the wire itself, disconnecting it will make no difference surely?



Glad you got this sorted.
SR Smith 1

"If the fault was with the wire itself, disconnecting it will make no difference surely?"

Correct, as Roger disconnected it at the solenoid. As it started and ran then the fault is proved to the solenoid.
Paul Hunt

And the reason for the short still worries me as does the reason why the car will start without the white and light green line from the solenoid. Is the line there as a belt and braces answer to a particular set of circumstances ?
Roger Walker

The white/green from the starter is there to bypass the in built resistor, but only when the vehicle is cranking.


It'll still start as the coil draws current via the white/pink resisted wire in the harness. But obviously the voltage at the coil + is lower than if the system is working correctly.
SR Smith 1

You said right at the beginning of this thread that you had 12v at the coil ignition on. It probably means the resistor wire has been by-passed. But check which coil you have, also as described earlier. If it is a ballast coil, continuous full battery voltage will damage it and leave you motionless!!
Allan Reeling

Allan. Many thanks. Alarm bells are ringing.
R
Roger Walker

"You said right at the beginning of this thread that you had 12v at the coil ignition on. It probably means the resistor wire has been by-passed. "

Not necessarily. As mentioned earlier you will only see less than 12v at the coil +ve of a ballasted system when it is passing current, and it will only be doing that if the points or electronic trigger are 'closed'. It's true that when switching off a running engine points at least are usually closed as the engine usually stops in one of two positions. But with some electronic triggers they will only 'close' and energise the coil when they detect cranking. When you have cranked an engine but it hasn't started, it may stop in a completely different position where even the points are open.

"It probably means the resistor wire has been by-passed."

If that were the case, then with this fault of the solenoid somehow earthing the white/light-green instead of connecting 12v to it, then full ignition voltage would have been shorted to earth while cranking, which will burn the wiring back through the white circuit and the ignition switch.

It's easy to check whether the ballast is connected or not. Remove the trigger wires from the coil, leaving just the harness white/light-greens. You should have 12v on the coil +ve with the ignition on. Now connect an earth to the coil -ve, and now the coil +ve should show about 6v. If it still shows 12v, then the ballast has indeed been bypassed.

In both cases you should also measure the primary resistance of the coil i.e. between the spades. If you see about 1.5 ohms that is a 6v coil and so needs ballasted ignition i.e. showing 6v on the coil +ve with an earth on the -ve. If you see about 3 ohms then it is a 12v coil and needs 12v on the +ve when there is an earth on the -ve.

You need to match the coil to the ignition wiring, or vice-versa. If you put a 12v coil on a ballasted system you will get a weak spark. If you put a 6v coil on a non-ballasted system, then this rapidly burns the points, or may damage an electronic trigger. I've not personally heard of the coil itself being damaged, even though it will get much hotter - possibly because the points can fail before that happens!
Paul Hunt

Paul, Useful advice - thanks
R
Roger Walker

Well the starter works but the solenoid definitely has a short. The problem now is to convince MGOC spares that it should be replaced under guarantee. Their initial position is "the starter is good enough".
R
Roger Walker

"the starter is good enough"

Hardly :o)
Paul Hunt

"the starter is good enough".
If it's not operating as intended then it's "not fit for purpose".
Sale of goods act comes in there.
Get insistent.
Allan Reeling

Paul, Allan, we are singing from the same hymnsheet.
R
Roger Walker

Spoke to Roger P at the MGOC who agreed with us. The motor will be replaced on Monday.
Again - thanks for all your advice and moral support.
R
Roger Walker

They can need some nudging, it took me several goes before they replaced a brake light switch that failed in less than a year. The replacement for that got a relay!
Paul Hunt

I had quite a struggle with MGOC when they sold me a clutch master cylinder that simply didn't fit. Even when I emailed them photos showing how it didn't fit, they still couldn't (or wouldn't) believe it. What happened to "the customer is always right"? I've never used them since and haven't been a member for years.
Mike Howlett

This thread was discussed between 05/03/2016 and 02/04/2016

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.